search_openalex.py

#!/usr/bin/env python3
"""Comprehensive OpenAlex search for NeKo-PIGNN literature review."""
import urllib.request
import urllib.parse
import json
import time

def search_openalex(query, per_page=25):
    """Search OpenAlex API."""
    url = f'https://api.openalex.org/works?filter=title_and_abstract.search:{urllib.parse.quote(query)}&sort=publication_date:desc&per_page={per_page}'
    req = urllib.request.Request(url, headers={'User-Agent': 'Mozilla/5.0 (research bot)'})
    with urllib.request.urlopen(req, timeout=30) as resp:
        return json.loads(resp.read().decode('utf-8'))

def extract_papers(data, min_year=2024):
    """Extract relevant paper info from OpenAlex response."""
    papers = []
    for r in data.get('results', []):
        year = r.get('publication_year', 0)
        if year < min_year:
            continue
        title = r.get('title', 'No title')
        date = r.get('publication_date', str(year))
        doi = r.get('doi', '')
        authors = []
        for a in r.get('authorships', []):
            author = a.get('author', {})
            authors.append(author.get('display_name', '?'))
        loc = r.get('primary_location', {})
        source = loc.get('source', {})
        journal = source.get('display_name', '')
        is_oa = r.get('open_access', {}).get('is_oa', False)
        cited_by = r.get('cited_by_count', 0)
        papers.append({
            'title': title,
            'date': date,
            'year': year,
            'doi': doi,
            'authors': authors,
            'journal': journal,
            'is_oa': is_oa,
            'cited_by': cited_by,
            'url': loc.get('landing_page_url', '')
        })
    return papers

def print_papers(papers, label):
    print(f"\n{'='*80}")
    print(f"📚 {label} ({len(papers)} papers)")
    print(f"{'='*80}")
    if not papers:
        print("  (no results)")
        return
    for i, p in enumerate(papers, 1):
        print(f"\n  [{i}] [{p['year']}-{p['date'][5:10] if p['date'] else '??'}] {p['title'][:140]}")
        print(f"      Journal: {p['journal'][:60]}")
        print(f"      Authors: {', '.join(p['authors'][:4])}{' et al.' if len(p['authors']) > 4 else ''}")
        print(f"      DOI: {p['doi']}")
        print(f"      Cited: {p['cited_by']} | OA: {p['is_oa']}")

QUERIES = [
    # Core queries for NeKo-PIGNN literature
    ("neural+koopman+operator+wildfire+propagation", "Neural Koopman + Wildfire Propagation"),
    ("physics+informed+graph+neural+network+wildfire+spread", "PI-GNN + Wildfire Spread"),
    ("koopman+operator+graph+neural+network+dynamics", "Koopman + GNN + Dynamics"),
    ("digital+twin+wildfire+deep+learning+prediction", "Digital Twin Wildfire DL"),
    ("GOES+ABI+wildfire+detection+deep+learning+2025", "GOES ABI Fire Detection 2025+"),
    ("wildfire+propagation+graph+attention+network+2025", "Graph Attention Wildfire 2025+"),
    ("unsupervised+wildfire+detection+satellite+imagery", "Unsupervised Wildfire Satellite Detection"),
    ("Rothermel+deep+learning+neural+network", "Rothermel + Deep Learning"),
    ("wildfire+caatinga+cerrado+machine+learning", "Wildfire Caatinga Cerrado ML"),
    ("brazil+wildfire+deep+learning+satellite+detection+2025", "Brazil Wildfire DL Satellite"),
    ("neural+koopman+physics+informed+neural+network", "Neural Koopman + PINN"),
    ("wildfire+spread+neural+ode+graph", "Wildfire + Neural ODE + Graph"),
    ("hybrid+physics+data+driven+wildfire+propagation+model", "Hybrid Physics-ML Wildfire"),
    ("GOES-19+satellite+fire+detection+thermal", "GOES-19 Fire Detection"),
    ("wildfire+edge+computing+real+time+detection", "Edge AI Real-time Wildfire"),
    ("knowledge+distillation+wildfire+detection+satellite", "Knowledge Distillation Wildfire"),
    ("explainable+AI+wildfire+prediction+spread", "XAI Wildfire Prediction"),
]

for query, label in QUERIES:
    try:
        data = search_openalex(query, per_page=15)
        papers = extract_papers(data, min_year=2024)
        # Print only 2025-2026 papers
        recent = [p for p in papers if p['year'] >= 2025]
        print_papers(recent if recent else papers[:5], label)
        time.sleep(0.3)
    except Exception as e:
        print(f"\n⚠️ Error for '{label}': {e}")